The ability to predict how changes in genotype affect phenotype would have wide applications, perhaps most importantly the rational identification of genes whose mutation predisposes an individual to a given disease. In this grant, we propose to develop computational approaches for predicting the phenotypic outcome of genetic perturbations in humans, and to apply these approaches to discover genes associated with polygenic diseases. We focus in particular on discovering genes underlying neural tube defects (NTDs), the second most common cause of human birth defects affecting nearly 1 in 1000 live births world-wide, and the most common permanently disabling birth defect in the United States. Given the extremely polygenic nature of these diseases, it is critical that systems biology-based approaches such as those proposed here be applied to understand their complex genetic basis. Each specific aim involves development or validation of a distinct approach for associating genes with diseases, in particular neural tube birth defects, with Specific Aim 1 on the application of large-scale gene networks to associate genes with diseases, identifying specific candidate human genes likely to play roles in neural tube defects, Specific Aim 2 on applying comparative evolutionary principles to identify appropriate, non-obvious, animal models of diseases, using this strategy to identify additional candidate neural tube defect genes, and Specific Aim 3 focused on the experimental testing in the model vertebrate animal Xenopus laevis of genes deemed most likely to be associated with human neural tube defects, using first studies of expression of at least 200 genes, followed by knockdown of each of up to 40 genes, validating and characterizing resulting neural tube defects. We will then search for genetic variation in the top roughly 300 candidate genes in DNA from biopsies of human infants with spina bifida, thereby confirming genes' involvement in human NTDs and identifying specific variants underlying neural tube defects. This work will increase our understanding of the genetic basis of neural tube birth defects and will be a step towards developing genetic diagnostics for susceptibility to these debilitating diseases. Success of these aims will also provide new theoretical methods for linking genes to diseases, a new theoretical framework for comparative analyses of diseases and mutational phenotypes across organisms, may suggest appropriate and new models for selected human diseases, and will provide mechanistic insights into the genetic basis of polygenic diseases in general and of neural tube defect diseases including spina bifida in particular.